重型AT锁止阀的流量特性分析与优化

为了提高锁止阀开启中液动力的计算效率,通过动网格技术在流体动力学软件Fluent中模拟锁止阀的阀芯开启过程,在对阀口过流面积进行数值计算的基础上,分析其流量及流量系数随阀芯位移的变化情况。使用BP神经网络对随机抽样法获得的样本数据进行训练和预测,并对预测的结果进行准确性评估。最后,以BP神经网络预测的数学模型为基础,使用遗传算法对锁止阀节流槽参数进行优化。结果表明：优化后的锁止阀开启过程中流量-开度特性更加稳定,有利于提高锁止阀的工作性能。     

7500T大型压铸机增压阀块内流道流场仿真分析及优化

设计性能优良的阀块对提升液压系统性能具有重要意义。利用计算动力学方法对7500T大型压铸机增压阀块的流阻特性进行分析,计算得到两种不同压铸机增压阀块结构在不同流量时的压力损失。结果表明：增压阀块压力损失随着流量的增大呈现出近似线性增大的变化情况;加设倾斜流道的压力损失比没有倾斜流道的压力损失小,且减阻效果随着流量的增大而增大。利用正交试验建立了4因素5水平的正交试验数据库,对斜流道增压阀进行多参数单目标优化,然后基于遗传算法协同BP神经网络优化斜流道增压阀块之后,可使增压阀块的压力损失下降20%左右。     

基于遗传算法的多级主压阀动态特性优化

多级主调压阀作为自动变速器液压控制系统的关键元件,其动态特性直接影响系统的快速性、稳定性和控制精度。在对阀进行结构参数优化时,先凭借经验对参数逐个调整再通过试验验证的方法,效率低且不能保证阀的性能最优。根据多级主压阀工作原理建立了多级主压阀数学模型;搭建多级主压阀Simulink仿真模型,通过分析多级主调压阀各结构参数对阀入口压力动态响应特性的影响,选取弹簧刚度、弹簧预压缩量和阻尼孔直径作为待优化变量;利用MATLAB优化工具箱中提供的遗传算法与Simulink结合对多级主调压阀的动态性能进行优化,并搭建试验台进行试验验证。仿真和试验结果表明：优化后多级主调压阀静态特性良好,负载流量突增时主压压力降低现象得到了改善,响应时间短,动态性能好。     

基于人工神经网络与遗传算法的Al-Mg-Si系合金抗拉强度预测模型

为了研究Al-Mg-Si系合金热处理制度和合金成分对力学性能的影响规律,采用人工神经网络（artificial neural network, ANN）和遗传算法（genetic algorithm, GA）相结合的方法,构建了Al-Mg-Si系合金强度预测模型（ANN-GA模型）。通过单因素和双因素分析,研究了合金元素含量和热处理工艺参数对铝合金抗拉强度的影响规律。结果表明,随着Si含量的增加,铝合金的抗拉强度呈现先降低后升高的趋势;随着Mg含量的增加、Cu含量的增加或者Fe含量的减少,铝合金的抗拉强度整体上呈现升高的趋势。双因素分析更能反映输入参数对铝合金抗拉强度的影响。Mg/Si比、Mg+Si总量和时效时间对Al-Mg-Si系合金力学性能的影响显著。铝合金的硬度随时间的变化趋势与ANN-GA模型的计算结果一致,峰值时效时间为29 h,相对误差为11.86%。     

一种基于遗传算法的钣金零件排样算法

钣金类零件是通过压力成形技术制造的一种应用广泛的零件,现今钣金件的数字化制造手段日益广泛。钣金工艺设计是钣金设计和制造之间的重要连接环节,科学合理的工艺是生产优良产品的重要因素。但钣金产品的工艺设计水平却远远落后于其设计和制造水平,成为了制约产品质量和成品的瓶颈。本文以一种复杂的锂电池箱体为例,对钣金件的数字化制造进行研究。首先提出了钣金件数字化集成制造的总体流程,研究了钣金件的设计、展开、板材的排样、激光切割、数控冲压和折弯等技术,实现了锂电池钣金箱体的集成制造。     

针对表面粗糙度和刀具振幅的切削用量优化研究

对进给量、切削速度和轴向切深这3个切削参数对工件表面粗糙度和刀具振动幅度的影响进行试验研究。采用BBD响应面法对6061铝工件进行端铣加工试验,并通过数学建模对试验结果进行分析。提出一种基于遗传算法的多目标优化方法来同时减小工件表面粗糙度和刀具振动幅度。建立能预报表面粗糙度和刀具振动的径向基神经网络模型,并通过试验验证其准确性。     

Optimization of variable demand fuzzy economic order quantity inventory models without and with backordering

A variable demand inventory model was developed for minimizing inventory cost, treating the holding and ordering costs and demand as independent fuzzy variables. Thereafter, backordering cost was also considered as an independent fuzzy variable. Fuzzy expected value model and fuzzy dependent chance programming model were constructed to find the optimal economic order quantity, which would minimize the fuzzy expected value of the total cost, so that the credibility of the total cost not exceeding a certain budget level was maximized. Optimization was carried out using genetic algorithms and particle swarm optimization algorithm, and their performances were compared. The developed model was found to be efficient not only in one artificial case study but also in two data sets collected from the industries. Therefore, this model could solve real-world problems, too.     

Fault diagnosis of rolling bearings using a genetic algorithm optimized neural network

In rotary complex machines, collapse of a component may inexplicably occur usually accompanied by a noise or a disturbance emanating from other sources. Rolling bearings constitute a vital part in many rotational machines and the vibration generated by a faulty bearing easily affects the neighboring components. Continuous monitoring, fault diagnosis and predictive maintenance, is a crucial task to reduce the degree of damage and stopping time for a rotating machine. Analysis of fault-related vibration signal is a usual method for accurate diagnosis. Among the resonant demodulation techniques, a well-known resolution often used for fault diagnosis is envelope analysis. But, usually this method may not be adequate enough to indicate satisfactory results. It may require some auxiliary additional techniques. This study suggests some methods to extract features using envelope analysis accompanied by Hilbert Transform and Fast Fourier Transform. The proposed artificial neural network (ANN) based fault estimation algorithm was verified with experimental tests and promising results. Every test was initiated with a reference ANN architecture to avoid inappropriate classification during the evaluation of fitness value. Later, ANN model was modified using a genetic algorithm providing, an optimal skillful fast-reacting network architecture with improved classification results.     

Analysis of the sensor placement for optimal temperature distribution reconstruction

The temperature distribution as a function of time and space is reconstructed over a non-homogeneous media having an arbitrary three-dimensional geometry. This is done by applying an inverse problem to the collected data from optimally placed sensors on the boundary surface of the object under study. Sensor positioning and the choice of the number of sensors are optimized in terms of the resolution of the reconstructed temperature field and the error propagation of the method in case of uncertain measurements. The method can be performed in real time since the major computation burden is performed off-line.     

Optimal conditions of isobutane dehydrogenation in radial flow moving bed hydrogen-permselective membrane reactors to enhance isobutene and hydrogen production

In the isobutane dehydrogenation process, coupling reaction and separation and optimization of the intensified process can improve the isobutane conversion and selectivity, reduce operational costs and lets to produce pure hydrogen. In this research, the radial flow moving bed reactors in the Olefex technology have been supported by Pd–Ag membrane plate to remove hydrogen from the reaction zone. The reactions occur in the tube side and the hydrogen is permeated from the reaction zone to the sweep gas stream. The proposed configuration has been modeled heterogeneously based on the mass and energy conservation laws considering reaction networks. To prove the accuracy of the considered model, the simulation results of the conventional process have been compared against available plant data. The Genetic algorithm as an effective method in the global optimization has been considered to optimize the operating condition of membrane reactors to enhance isobutene productivity. In this optimal configuration, the isobutene production has been enhanced about 3.7%.